[BC-breaking] Remove _new_empty_tensor op

test/test_onnx.py

@@ -76,20 +76,6 @@ def to_numpy(tensor):
                 else:
                     raise
 
-    @unittest.skip("Disable test until Split w/ zero sizes is implemented in ORT")
-    def test_new_empty_tensor(self):
-        class Module(torch.nn.Module):
-            def __init__(self):
-                super(Module, self).__init__()
-                self.conv2 = ops.misc.ConvTranspose2d(16, 33, (3, 5))
-
-            def forward(self, input2):
-                return self.conv2(input2)
-
-        input = torch.rand(0, 16, 10, 10)
-        test_input = torch.rand(0, 16, 20, 20)
-        self.run_model(Module(), [(input, ), (test_input,)], do_constant_folding=False)
-
     def test_nms(self):
         boxes = torch.rand(5, 4)
         boxes[:, 2:] += torch.rand(5, 2)

test/test_ops.py

@@ -449,15 +449,6 @@ def test_autocast(self):
                 self.test_nms_cuda(dtype=dtype)
 
 
-class NewEmptyTensorTester(unittest.TestCase):
-    def test_new_empty_tensor(self):
-        input = torch.tensor([2., 2.], requires_grad=True)
-        new_shape = [3, 3]
-        out = torch.ops.torchvision._new_empty_tensor_op(input, new_shape)
-        assert out.size() == torch.Size([3, 3])
-        assert out.dtype == input.dtype
-
-
 class DeformConvTester(OpTester, unittest.TestCase):
     def expected_fn(self, x, weight, offset, mask, bias, stride=1, padding=0, dilation=1):
         stride_h, stride_w = _pair(stride)

torchvision/ops/__init__.py

@@ -1,6 +1,5 @@
 from .boxes import nms, batched_nms, remove_small_boxes, clip_boxes_to_image, box_area, box_iou, generalized_box_iou
 from .boxes import box_convert
-from .new_empty_tensor import _new_empty_tensor
 from .deform_conv import deform_conv2d, DeformConv2d
 from .roi_align import roi_align, RoIAlign
 from .roi_pool import roi_pool, RoIPool
@@ -19,7 +18,7 @@
     'deform_conv2d', 'DeformConv2d', 'nms', 'batched_nms', 'remove_small_boxes',
     'clip_boxes_to_image', 'box_convert',
     'box_area', 'box_iou', 'generalized_box_iou', 'roi_align', 'RoIAlign', 'roi_pool',
-    'RoIPool', '_new_empty_tensor', 'ps_roi_align', 'PSRoIAlign', 'ps_roi_pool',
+    'RoIPool', 'ps_roi_align', 'PSRoIAlign', 'ps_roi_pool',
     'PSRoIPool', 'MultiScaleRoIAlign', 'FeaturePyramidNetwork',
     'sigmoid_focal_loss'
 ]

torchvision/ops/_register_onnx_ops.py

@@ -38,18 +38,7 @@ def roi_pool(g, input, rois, spatial_scale, pooled_height, pooled_width):
                         pooled_shape_i=(pooled_height, pooled_width), spatial_scale_f=spatial_scale)
         return roi_pool, None
 
-    @parse_args('v', 'is')
-    def new_empty_tensor_op(g, input, shape):
-        dtype = input.type().scalarType()
-        if dtype is None:
-            dtype = 'Float'
-        dtype = scalar_type_to_onnx.index(cast_pytorch_to_onnx[dtype])
-        shape = g.op("Constant", value_t=torch.tensor(shape))
-        return g.op("ConstantOfShape", shape,
-                    value_t=torch.tensor([0], dtype=scalar_type_to_pytorch_type[dtype]))
-
     from torch.onnx import register_custom_op_symbolic
     register_custom_op_symbolic('torchvision::nms', symbolic_multi_label_nms, _onnx_opset_version)
     register_custom_op_symbolic('torchvision::roi_align', roi_align, _onnx_opset_version)
     register_custom_op_symbolic('torchvision::roi_pool', roi_pool, _onnx_opset_version)
-    register_custom_op_symbolic('torchvision::_new_empty_tensor_op', new_empty_tensor_op, _onnx_opset_version)